PROJECT SUMMARY Treatment of patients with glioma has been largely unsuccessful, and even if initial treatment shows some effect, the long-term prognosis remains poor because recurrence of the disease is near-certain. A significant challenge in establishing treatments for recurrent gliomas is that the diffuse infiltrative growth and presumed extensive tumor heterogeneity allows tumor cells to ?escape? and even develop resistance to therapy. In order for better treatments to be developed we need to understand both the extent of tumor heterogeneity and how it evolves in response to treatment. Although such a need is easily stated it has been difficult, if not impossible, to achieve with current resources because it requires tracking and characterizing tumor changes within the same patient. Moreover, because of patient to patient variability, and the need for statistical power, these types of investigations also require standardized data from hundreds of patients. With these requirements in mind, and the long-term goal of identifying new, effective treatment targets, we initiated the international Glioma Longitudinal Analysis (GLASS) Consortium. Since 2014, GLASS has established the largest (and still expanding) database of tumor samples sequenced at different time points among any cancer type. Data from these samples (collected at 34 institutions in 12 countries) are now aggregated and integrated with patient and phenotypic data across international sub studies. The GLASS dataset is supported by an infrastructure that standardizes key parameters across studies/sites so that complex, integrated analyses are possible. Preliminary analysis of this unique dataset demonstrated that it will allow us to create a portrait of the recurrence process and discover novel molecular vulnerabilities that can be targeted for successful therapeutic intervention. We are now poised to further exploit the GLASS data to identify critical processes driving glioma evolution. To do this we propose: Aim1 - To test the hypothesis that clonal diversity (tumor heterogeneity) is significantly impacted by treatment, and Aim 2 - To test the hypothesis that immunoediting results in the selection of glioma cells that are capable of evading the immune response. Upon completion of these aims, we will have gained new insights into how treatment and the immune system drive the clonal (tumor cell) selection that leads to glioma tumor heterogeneity. In the process, we will also establish and share the tools/approaches needed for valid analyses of this type of multi-dimensional, multi-time point data. Taken together, the results of these efforts should identify novel avenues for treatment with better, more reliable outcomes.